[utils] fix lit -vv alias, remove rendundant uses of '-a -v'

-vv should be the same as -v

A bunch of lit tests use both -a and -v (or -vv), which was always
redundant, but now results in the last option being applied, rather than
-a always overriding -v.

Author: hnrklssn

llvm/utils/lit/lit/cl_arguments.py

@@ -189,7 +189,7 @@ def parse_args():
         action=AliasAction,
         alias=[
             lambda namespace: TestOutputAction.setOutputLevel(
-                namespace, "test_output", "all"
+                namespace, "test_output", "failed"
             ),
         ],
     )

llvm/utils/lit/tests/per-test-coverage-by-lit-cfg.py

@@ -1,10 +1,10 @@
 # Test if lit_config.per_test_coverage in lit.cfg sets individual test case coverage.
 
-# RUN: %{lit} -a -vv -Dexecute_external=False \
+# RUN: %{lit} -a -Dexecute_external=False \
 # RUN:     %{inputs}/per-test-coverage-by-lit-cfg/per-test-coverage-by-lit-cfg.py | \
 # RUN:   FileCheck -DOUT=stdout %s
 
-# RUN: %{lit} -a -vv -Dexecute_external=True \
+# RUN: %{lit} -a -Dexecute_external=True \
 # RUN:     %{inputs}/per-test-coverage-by-lit-cfg/per-test-coverage-by-lit-cfg.py | \
 # RUN:   FileCheck -DOUT=stderr %s
 

llvm/utils/lit/tests/per-test-coverage.py

@@ -1,10 +1,10 @@
 # Test LLVM_PROFILE_FILE is set when --per-test-coverage is passed to command line.
 
-# RUN: %{lit} -a -vv --per-test-coverage -Dexecute_external=False \
+# RUN: %{lit} -a --per-test-coverage -Dexecute_external=False \
 # RUN:     %{inputs}/per-test-coverage/per-test-coverage.py | \
 # RUN:   FileCheck -DOUT=stdout %s
 
-# RUN: %{lit} -a -vv --per-test-coverage -Dexecute_external=True \
+# RUN: %{lit} -a --per-test-coverage -Dexecute_external=True \
 # RUN:        %{inputs}/per-test-coverage/per-test-coverage.py | \
 # RUN:   FileCheck -DOUT=stderr %s
 

llvm/utils/lit/tests/shtest-cat.py

@@ -1,6 +1,6 @@
 ## Test the cat command.
 #
-# RUN: not %{lit} -a -v %{inputs}/shtest-cat \
+# RUN: not %{lit} -v %{inputs}/shtest-cat \
 # RUN: | FileCheck -match-full-lines %s
 # END.
 

llvm/utils/lit/tests/shtest-env-negative.py

@@ -1,6 +1,6 @@
 ## Test the env command (failing tests).
 
-# RUN: not %{lit} -a -v %{inputs}/shtest-env-negative \
+# RUN: not %{lit} -v %{inputs}/shtest-env-negative \
 # RUN: | FileCheck -match-full-lines %s
 #
 # END.

llvm/utils/lit/tests/shtest-env-path.py

@@ -1,9 +1,9 @@
 ## Tests env command for setting the PATH variable.
 
 # The test is using /bin/sh. Limit to system known to have /bin/sh.
-# REQUIRES: system-linux
+# REQUIRES: system-linux || system-darwin
 
-# RUN: %{lit} -a -v %{inputs}/shtest-env-path/path.txt \
+# RUN: %{lit} -a %{inputs}/shtest-env-path/path.txt \
 # RUN:   | FileCheck -match-full-lines %s
 #
 # END.

llvm/utils/lit/tests/shtest-export.py

@@ -1,6 +1,6 @@
 ## Test the export command.
 
-# RUN: not %{lit} -a -v %{inputs}/shtest-export \
+# RUN: not %{lit} -v %{inputs}/shtest-export \
 # RUN: | FileCheck -match-full-lines %s
 #
 # END.

llvm/utils/lit/tests/shtest-glob.py

@@ -1,6 +1,6 @@
 ## Tests glob pattern handling in echo command.
 
-# RUN: not %{lit} -a -v %{inputs}/shtest-glob \
+# RUN: not %{lit} -v %{inputs}/shtest-glob \
 # RUN: | FileCheck -dump-input=fail -match-full-lines --implicit-check-not=Error: %s
 # END.
 

llvm/utils/lit/tests/shtest-pushd-popd.py

@@ -1,6 +1,6 @@
 # Check the pushd and popd commands
 
-# RUN: not %{lit} -a -v %{inputs}/shtest-pushd-popd \
+# RUN: not %{lit} -v %{inputs}/shtest-pushd-popd \
 # RUN: | FileCheck -match-full-lines %s
 #
 # END.

llvm/utils/lit/tests/shtest-readfile-external.py

@@ -4,7 +4,7 @@
 # ALLOW_RETRIES: 2
 
 # UNSUPPORTED: system-windows
-# RUN: env LIT_USE_INTERNAL_SHELL=0 not %{lit} -a -v %{inputs}/shtest-readfile | FileCheck -match-full-lines -DTEMP_PATH=%S/Inputs/shtest-readfile/Output %s
+# RUN: env LIT_USE_INTERNAL_SHELL=0 not %{lit} -v %{inputs}/shtest-readfile | FileCheck -match-full-lines -DTEMP_PATH=%S/Inputs/shtest-readfile/Output %s
 
 # CHECK: -- Testing: 5 tests{{.*}}